feat(product): 聚水潭销售主题分析按日采集工作流

collect_erp_sales_theme_daily.py 采集 bi.erp321.com「报表→销售主题分析→按日统计」，
异步导出 + OSS 直连兜底，固定品牌+店铺口径，写 erp_sales_theme_daily_metrics
（db 层建表与 upsert 同步，每日一行幂等覆盖）。注册 product.erp_sales_theme_daily.collect
手动命令，复用 erp-account 登录绑定新增 CDP 22167；目录/集成计数 60→61，
新脚本纳入 CI 聚焦 lint 清单。

tests/modules/product_commerce/test_erp_sales_theme_daily.py

@@ -0,0 +1,191 @@
+# -*- coding: utf-8 -*-
+"""聚水潭销售主题分析(按日)采集脚本的纯解析与 CLI 参数测试。"""
+from __future__ import annotations
+
+import sys
+from pathlib import Path
+
+import pytest
+
+from gyxx_flow.modules import product_commerce
+from gyxx_flow.modules.product_commerce import (
+    collect_erp_sales_theme_daily as collector,
+)
+from gyxx_flow.modules.product_commerce import db as product_db
+
+HEADER = [
+    "日期", "销售单数", "销售数量", "实发数量", "实发金额", "销售金额",
+    "退货单数", "退货数量", "实退数量", "退货金额", "实退金额",
+    "净销售单数", "净销量", "净销售额", "运费收入", "运费支出", "实发单数",
+]
+
+
+def _write_workbook(path: Path, rows: list[list]) -> None:
+    import openpyxl
+
+    workbook = openpyxl.Workbook()
+    sheet = workbook.active
+    for row in rows:
+        sheet.append(row)
+    workbook.save(str(path))
+    workbook.close()
+
+
+def test_parse_sales_theme_workbook_roundtrip(tmp_path: Path) -> None:
+    xlsx = tmp_path / "销售主题分析_按日_1.xlsx"
+    _write_workbook(
+        xlsx,
+        [
+            HEADER,
+            ["2026-10-11", "649", 655, 646, "173,129.23", 175752.83,
+             15, 15, 0, 2677.47, "0.00", 634, 640, 173075.36, "20.00", None, 640],
+            ["2026-10-10", 690, 691, 685, 166761.58, 168185.97,
+             9, 9, 0, 2656.71, 0, 681, 686, 166529.13, None, 1.6, 685],
+            ["##", 1339, 1346, 1331, 339890.81, 343938.8,
+             24, 24, 0, 5334.18, 0, 1315, 1326, 339604.49, 20, 1.6, 1325],
+        ],
+    )
+
+    records = collector.parse_sales_theme_workbook(xlsx)
+
+    assert [record["stat_date"] for record in records] == ["2026-10-11", "2026-10-10"]
+    first = records[0]
+    assert first["sales_order_count"] == 649
+    assert first["sales_qty"] == 655
+    assert first["shipped_amount"] == 173129.23
+    assert first["sales_amount"] == 175752.83
+    assert first["return_amount"] == 2677.47
+    assert first["returned_amount"] == 0.0
+    assert first["net_sales_order_count"] == 634
+    assert first["net_sales_amount"] == 173075.36
+    assert first["freight_income"] == 20.0
+    assert first["freight_expense"] is None
+    assert first["shipped_order_count"] == 640
+    # 合计行(##)必须被跳过
+    assert len(records) == 2
+
+
+def test_parse_sales_theme_workbook_rejects_missing_date_column(tmp_path: Path) -> None:
+    xlsx = tmp_path / "bad.xlsx"
+    _write_workbook(xlsx, [["销售单数", "销售数量"], [1, 2]])
+    with pytest.raises(ValueError):
+        collector.parse_sales_theme_workbook(xlsx)
+
+
+def test_parse_number_and_amount_helpers() -> None:
+    assert collector.parse_number("1,339") == 1339
+    assert collector.parse_number(None) == 0
+    assert collector.parse_number("") == 0
+    assert collector.parse_amount(None) is None
+    assert collector.parse_amount("") is None
+    assert collector.parse_amount("1.6") == 1.6
+    assert collector.parse_amount(2677.467) == 2677.47
+
+
+class _CaptureCursor:
+    def __init__(self) -> None:
+        self.batches: list[tuple] = []
+
+    def __enter__(self):
+        return self
+
+    def __exit__(self, *exc_info):
+        return False
+
+
+class _CaptureConnection:
+    def __init__(self) -> None:
+        self.cursor_obj = _CaptureCursor()
+
+    def cursor(self):
+        return self.cursor_obj
+
+
+def test_upsert_sales_theme_dedups_and_normalizes(monkeypatch: pytest.MonkeyPatch) -> None:
+    captured: dict[str, object] = {}
+
+    def fake_execute_values(cur, sql, rows, template=None):
+        captured["sql"] = sql
+        captured["rows"] = list(rows)
+
+    monkeypatch.setattr(product_db, "execute_values", fake_execute_values)
+    records = [
+        {
+            "stat_date": "2026-10-11",
+            "sales_order_count": 649,
+            "freight_income": "20.00",
+        },
+        {
+            # 同日重复: 整体覆盖前一条(同 upsert_erp_all_shop 语义), 数值缺省为 0
+            "stat_date": "2026-10-11",
+            "sales_order_count": 999,
+            "brands": ["光影", "光影行星", "光影"],
+            "shops": ["天猫-光影行星旗舰店", "JD光影行星旗舰店"],
+            "source_file": "var/raw/erp销售主题分析日报/x.xlsx",
+        },
+        {
+            "stat_date": "2026-10-10",
+            "sales_qty": "691",
+        },
+    ]
+
+    conn = _CaptureConnection()
+    written = product_db.upsert_erp_sales_theme_daily_metrics(conn, records)
+
+    assert written == 2
+    rows = captured["rows"]
+    assert len(rows) == 2
+    by_date = {row[0]: row for row in rows}
+    # 同日重复记录取最后一条, 数字字段缺省为 0, 金额字符串转 float
+    assert by_date["2026-10-11"][1] == 999
+    assert by_date["2026-10-10"][2] == 691
+    tenth = by_date["2026-10-10"]
+    # NOT NULL 金额列缺省(导出空单元格)落 0; 运费两列允许 NULL
+    assert tenth[9] == 0  # return_amount
+    assert tenth[15] is None  # freight_income
+    assert tenth[16] is None  # freight_expense
+    eleventh = by_date["2026-10-11"]
+    assert eleventh[-4] == "发货时间"  # date_basis 默认
+    assert eleventh[-3] == ["光影", "光影行星"]  # brands 去重排序
+    assert eleventh[-2] == ["JD光影行星旗舰店", "天猫-光影行星旗舰店"]
+    assert eleventh[-1] == "var/raw/erp销售主题分析日报/x.xlsx"
+    # upsert 目标表与主键在 SQL 里声明
+    sql = str(captured["sql"])
+    assert "erp_sales_theme_daily_metrics" in sql
+    assert "ON CONFLICT (stat_date)" in sql
+
+
+def test_business_scope_constants_are_consistent() -> None:
+    assert len(collector.BRANDS) == 3
+    assert sorted(collector.BRANDS) == ["gyxx", "光影", "光影行星"]
+    assert len(collector.SHOP_NAMES) == len(set(collector.SHOP_NAMES)) > 0
+    for name in collector.SHOP_NAMES:
+        assert name == name.strip()
+
+
+def test_resolve_window_variants() -> None:
+    args = collector.parse_args(["--date", "2026-10-11"])
+    start, end = collector.resolve_window(args)
+    assert (start.isoformat(), end.isoformat()) == ("2026-10-11", "2026-10-11")
+
+    args = collector.parse_args(["--date", "2026-10-11", "--days", "7"])
+    start, end = collector.resolve_window(args)
+    assert (start.isoformat(), end.isoformat()) == ("2026-10-05", "2026-10-11")
+
+    args = collector.parse_args(["--from", "2025-06-01", "--to", "2026-01-01"])
+    start, end = collector.resolve_window(args)
+    assert (start.isoformat(), end.isoformat()) == ("2025-06-01", "2026-01-01")
+
+    with pytest.raises(SystemExit):
+        collector.resolve_window(collector.parse_args(["--from", "2026-01-01"]))
+    with pytest.raises(SystemExit):
+        collector.resolve_window(
+            collector.parse_args(["--from", "2026-01-01", "--to", "2025-01-01"])
+        )
+
+
+def test_module_uses_shared_cookie_skip_exit_code() -> None:
+    from gyxx_flow.core.exit_codes import COOKIE_SKIP_EXIT_CODE
+
+    assert COOKIE_SKIP_EXIT_CODE == 75
+    assert sys.modules[product_commerce.__name__] is not None

tests/test_runtime_integrations.py

@@ -102,10 +102,10 @@ def test_project_registry_covers_every_script_with_unique_stable_port() -> None:
         data_root=project_root / "var",
     )
 
-    assert len(scripts.command_ids) == 60
-    assert len(first.command_ids) == 157
+    assert len(scripts.command_ids) == 61
+    assert len(first.command_ids) == 158
     bindings = [first.binding_for(command_id) for command_id in first.command_ids]
-    assert len({binding.cdp_port for binding in bindings}) == 157
+    assert len({binding.cdp_port for binding in bindings}) == 158
     assert all(22000 <= binding.cdp_port <= 22999 for binding in bindings)
     assert all(binding.cdp_url.startswith("http://127.0.0.1:") for binding in bindings)
     assert bindings == [second.binding_for(item.script_id) for item in bindings]

tests/test_script_catalog.py

@@ -83,7 +83,7 @@ def test_default_registry_covers_every_executable_workflow_entry() -> None:
     assert scheduled_workflows == 39
     assert manual_workflows == 0
     assert len(workflow_entries) == 55
-    assert len(catalog.scripts) == 60
+    assert len(catalog.scripts) == 61
     assert workflow_entries < set(catalog.script_ids)
     assert {
         "content.mapping.rebuild",
@@ -280,7 +280,7 @@ def test_cli_lists_and_dry_runs_registered_commands(tmp_path: Path) -> None:
 
     assert main(["scripts", "list", "--json"], settings=settings, stdout=output) == 0
     rows = json.loads(output.getvalue())
-    assert len(rows) == 60
+    assert len(rows) == 61
     assert {row["module"] for row in rows} == {
         "content_marketing",
         "product_commerce",